Project Summary Phosphorylation can regulate protein function, which is a cornerstone of normal tissue development and home- ostasis. However, kinases, the enzymes that catalyze protein phosphorylation, are often dysregulated in cancer. Recently, advances have been made to measure global phosphorylation within human patient tumor samples. The hope is that this data holds the key to identifying patient-speci?c targets in cancer therapy. Unfortunately, challenges exist in interpreting phosphorylation data and its re?ection of the underlying dysregulation of signaling networks. The goal of this project is to develop an algorithm that translates the measurements of phosphorylation in human samples to a prediction of kinase activity pro?les. The kinase activity pro?les could then be used to iden- tify new targets and classify tumor types. This goal will be achieved by: the development of graph-based score, based on predicted kinase-substrate relationships, interpretation of that score through statistical frameworks, and testing and improvement of the algorithms on available control and patient data.